Qualitative and Quantitative Analysis of Six Sigma in Service Organizations

Transcription

1 Chapter 10 Qualitative and Quantitative Analysis of Six Sigma in Service Organizations Ayon Chakraborty and Kay Chuan Tan Additional information is available at the end of the chapter 1. Introduction Quality management has long been established as an important strategy for achieving competitive advantage. The aim of the businesses may differ, but the importance of customers is a matter of common interest. The ability of the organizations to adapt to new customer requirements in a globalized market is of vital importance for long-term success. Traditional quality initiatives such as statistical quality control, zero defects, and total quality management, have been key initiatives for many years. In last two decades, Six Sigma evolved as a new quality management initiative and now many organizations are working towards its implementation. Six Sigma is a disciplined approach for improving manufacturing or service processes, based on defined metrics (Hahn et al., 1999). The strength of Six Sigma lies in its well defined framework involving methodology applying different tools and techniques (Goh, 2002). The Six Sigma journey started from Motorola in 1980s and spread its importance through adoptions by different high profile organizations such as General Electric (GE), Honeywell, Asea Brown Bovari (ABB), Lockheed-Martin, Polaroid, and Texas Instruments (Goh, 2002; Hahn et al., 1999). This initial success of Six Sigma has seen its implementation spreading in several other organizations mostly in mass-manufacturing sector (McAdam et al., 2005). These organizations adopted the systematic framework of Six Sigma through training and project management practices (Brady and Allen, 2006). The use of Six Sigma has been relatively high among many western organizations till now, see, for example, Inozu et al. (2006), Raisinghani et al. (2005), and Antony (2004b), but there exists a diversity of opinion among researchers regarding the actual benefits of Six Sigma. Literature explaining about the positive effects on financial performance can be found in e.g. Jones Jr. (2004), Goh (2002), Caulcutt (2001), and Rucker (2000). However, McAdam and Lafferty (2004), Senapati (2004), and Paul (1999), for instance, express a more pessimistic view regarding the benefit of Six Sigma investments Chakraborty and Tan, licensee InTech. This is an open access chapter distributed under the terms of the Creative Commons Attribution License ( which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

2 248 Total Quality Management and Six Sigma Similar to Six Sigma, services in the last two decades have become an important part in economies of developed as well as for developing nations. The importance of services has also increased as it became a major employment provider (Cook et al., 1999). This increased importance of service sector has various researchers contributing to the service literature. The service research from its beginning can be divided into stages such as an initial realization of the difference between goods and service, the development of conceptual frameworks, the empirical testing of these frameworks and the application of the tools and frameworks to improve service management (Johnston, 1999). The various stages of service research have gained by the major contributions from marketing and after that to some extent from operations management field. But in case of definitions, services still lack a unified definition and similar can be said about the classification scheme. So, there is a need to provide a universally accepted classification scheme which can be done through empirical derivation and considering different dimensions of service organizations. This will facilitate in exploration of service quality and service strategy (Cook et al., 1999). As service quality is now the major focus of service organizations, so a better understanding of unique characteristics of services will be helpful. This in turn will help spreading new quality initiatives such as Six Sigma, in services. This spread, however, is limited in service industries. A key argument here is that many service processes are unseen, intangible, and even immeasurable. As such, they are not amendable to improvement using a Six Sigma approach. This thinking has turned out to be rather presumptuous at least for the health care, banking, and call center services which have been able to apply Six Sigma (Hensley and Dobie, 2005). Other services such as education and hospitality are also beginning to see Six Sigma applications. The question of Six Sigma and its implementation and performance in service organizations has not before been under investigation. The literature includes many papers describing Six Sigma implementation in a variety of business types; however, very few of these papers report empirical research and include single case studies (Hendry and Nonthaleerak, 2005). Examples of non-manufacturing contexts discussed in the literature include healthcare and financial services, as well as in non-production internal functions within a manufacturing organization (Nonthaleerak and Hendry, 2008). The paper by Wyper and Harrison (2000), discuss Six Sigma implementation in non-manufacturing context and highlights the difficulties specific to that context. Does et al. (2002), present a comparison of eight Six Sigma projects in non-manufacturing processes with a theoretical manufacturing application in a case study company in the Netherlands. This paper addressed various problems, typical of non-manufacturing and also identified difficulties in tools application. They conclude that Six Sigma can be applicable in non-manufacturing contexts with minor adaptations. Given that, the research is based in a single case study setting, there are limits on the degree to which the conclusions can be generalized (Nonthaleerak and Hendry, 2008). McAdam and Lafferty (2004) conducted a survey in a single company on Six Sigma implementation issue from process and people perspective. They found low success of Six Sigma in nonmanufacturing areas.

3 Qualitative and Quantitative Analysis of Six Sigma in Service Organizations 249 The literature on discussion about Six Sigma in service organizations also concentrate about issues in implementation due to inherent differences between manufacturing and service. The possible reason being the manufacturing roots of Six Sigma like other quality management initiatives. Antony (2004a), Benedetto (2003), and Sehwall and De Yong (2003), argue for example that there are certain differences in Six Sigma implementation in services from manufacturing which acts as a barrier in Six Sigma implementation in service organizations. Failed implementation initiatives, especially as extensive as Six Sigma implementation, result in financial losses and potential resistance towards change among the actors involved. It is therefore of importance that the implementation strategies used are well adapted, see, e.g. Biolos (2002). Hence, the literature has conflicting evidence regarding the applicability of Six Sigma to non-manufacturing settings and therefore there is a need to investigate further this issue (Nonthaleerak and Hendry, 2008). The studies so far focused on Six Sigma implementation in non-manufacturing context at project level. The studies are mostly single case studies and descriptive in nature. The survey based studies are either pilot survey or focused on a single organization. Thus, there are a number of key research gaps in the literature, which our research aims to address. There is insufficient empirical evidence to verify and further explain the Six Sigma CSFs identified in service organizations. The existing difficulties in Six Sigma implementation in service organizations are not well understood. There is a scope to contribute to Six Sigma implementation in service organizations by enhancing the knowledge about tools and techniques usage. Our research will focus on individual Six Sigma projects in service organizations to fill the identified research gaps. 2. Background 2.1. Six sigma definitions and philosophy In 1924, Walter A. Shewart from Bell Telephone Laboratories, proposed the concept of using statistical charts to control the variables of products manufactured at Western Electric. This was the beginning of statistical quality control (Small, 1956). Dr. Shewart kept on with his efforts and applied the fundamentals of statistical quality control to industry. This lead to the modern attention to the use of statistical tools for the manufacture of products and process, originated prior to and during World War II, when the United States of America geared up to a massive build-up of machinery and arms to successfully conclude the war (Brady, 2005). The Western Electric manufacturing company is noteworthy during this time because it was the breeding ground for many quality leaders, not only Shewart but Joseph Juran, Edwards Deming and Kaoru Ishikawa all worked there at some time (Dimock, 1977). Two prominent individuals were Deming and Juran. Deming promoted the use of the plan-do-check-act (PDCA) cycle of continuous improvement. Later Juran introduced the concepts of project by project quality improvement. Any discussion on quality today will most likely cite at least one

4 250 Total Quality Management and Six Sigma from the group of Deming, Juran, Crosby, Feigenbaum, and Ishikawa, if not all. They certainly represent the preponderance of information about quality. Adding to this group, Bill Smith, Motorola Vice President and Senior Quality Assurance Manager, is widely regarded as the father of Six Sigma, Shina (2002). Because Six Sigma was built on previous quality methodologies, a list of the pioneers of the quality and their contribution is included in Table 1. According to Shina (2002) before, January 15, 1987, Six Sigma was solely a statistical term. Since then, the Six Sigma crusade, which began at Motorola, has spread to other companies which are continually striving for excellence. At Motorola, Six Sigma is defined as A quality improvement program with a goal of reducing the number of defects to as low as 3.4 parts per million opportunities or %. Six Sigma has a number of different meanings and interpretations (Henderson and Evans, 2000, pp 261). Its origin comes from statistics where sigma represents the amount of variation about a process average. From a business view of point, Six Sigma may be defined as A business strategy used to improve business profitability, to improve the effectiveness and efficiency of all operations to meet or exceed customer s needs and expectations (Kwak and Anbari, 2006, pp 709). Various other definitions include: Six Sigma is a formal methodology for measuring, analysing, improving, and then controlling or locking-in processes. This statistical approach reduces the occurrence of defects from a three sigma level or defects per million to a Six Sigma level or less than four defects per million (Bolze, 1998). Six Sigma is a comprehensive, statistics-based methodology that aims to achieve nothing less than perfection in every single company process and product (Paul, 1999). Six Sigma is a disciplined method of rigorous data gathering and robust statistical analysis to pinpoint sources of error and ways of eliminating them (Harry and Schroeder, 1999). Six Sigma as an information-driven methodology for reducing waste, increasing customer satisfaction, and improving processes, with a focus on financially measurable results (As defined by Minitab in Goh, 2002). Quality Gurus Contribution Philip B. Crosby Senior manager involvement; four absolutes of quality management; quality costs measurements W. Edwards Deming Plan-do-study-act; top management involvement; concentration on system improvement; constancy of purpose Armand V. Feigenbaum Total quality control/management; top management involvement Kauro Ishikawa Cause and effect diagram; company-wide quality control; Joseph M. Juran Top management involvement; quality trilogy; quality cost measurement Walter A. Shewart Assignable cause versus chance cause; control charts; plan-do-check-act; use of statistics for improvement Table 1. Pioneers of quality and their contribution to Six Sigma knowledge bank (adapted from Wortman, 2001)

5 Qualitative and Quantitative Analysis of Six Sigma in Service Organizations 251 The statistical focus of various Six Sigma definitions reflects its basic philosophy. Six Sigma is an operating philosophy that can be shared beneficially by everyone, including customers, shareholders, employees, and suppliers. Fundamentally, it is also a customer-focused methodology that drives out waste, raises levels of quality, and improves the financial performance of organizations to breakthrough levels (Chua, 2001). Six Sigma s target for perfection is to achieve no more than 3.4 defects, errors, or mistakes per million opportunities whether this involves the design and production of a product or a customer-oriented service process. It is from this target that the name Six Sigma originated. Compared to a process that has greater variation, a process with less variation will be able to fit more standard deviations or sigmas between the process centre and its specification limits. An increase in the number of sigmas between the specification limits means the acceptance of fewer defects. More sigmas imply a more consistent manufacturing or service delivery process (Chua, 2001) Tools and techniques and six sigma methodologies The concept of Six Sigma was introduced at and popularized by Motorola in Six Sigma is a logical extension of Statistical Process Control (SPC). The concept behind SPC is simple enough but powerful, indeed. Variation is present in every production/operations process and such variation is due either to common causes or special causes. The breakthrough made by Shewart was the statistical definition and measurement of variation, where variation within three-sigma limits was deemed to be random and produced by common causes, and variation outside of the three-sigma limits was produced by special causes, indicating a process problem (Shewart, 1931). The ±3σ process limits mean a defect rate of 2.7/1000 or 2,700/1,000,000 opportunities, if one ignores lateral shifts in the process, and the capability of the process is thus defined as the range of natural variation, that is, ±3σ, or Cpk = 6σ. Six-Sigma doubles the range of normal variation to ±6σ, and allows for a 1.5σ lateral shift in the process average. The result is a dramatic tightening of acceptable defect rate target to 3.4/1,000,000 opportunities. The basic elements of Six Sigma are not new. SPC, failure mode effect analysis, gage repeatability and reproducibility studies, and other tools and techniques, have been in use for some time. Six Sigma offers a framework that unites these basic quality tools and techniques with high-level management support. There is much literature available on tools and techniques used in Six Sigma. Tools are mostly referred to as having a clearly defined role but narrow in focus, whereas techniques have wider application and require specific skills, creativity, and training (Antony, 2006). Similar to CSFs, CTQs, and KPIs; there is limited literature which discuss about STTs specific to service organizations. Discussion on STTs in the literature is mostly on its usage at different phases of DMAIC methodology. De Koning and De Mast (2006) used seven different literature sources and provided a summary of STTs used in DMAIC phases. Some other literature provide classification scheme for tools and techniques used. Henderson and Evans (2000) discussed about tool sets in three groups; team tools, process tools, and statistical tools. As for Six Sigma tools and techniques specific to service organizations, Antony (2006) provides a grid as a guideline for services.

6 252 Total Quality Management and Six Sigma A number of classification schemes for STTs exists, the majority of which are based on the DMAIC methodology. The classification schemes by the American Society for Quality (ASQ) and by Nancy Tague (1995) called the Tool Matrix provide an exhaustive list of tools and techniques which can be used during Six Sigma implementation. The ASQ classification scheme and the tool matrix have almost similar categories. The only difference being in the number of tools and techniques each category Classification scheme of tools and techniques a. ASQ Classification According to ASQ, tools and techniques that are utilized in different phases of DMAIC are classified according to their uses. There are 7 broad categories: Cause Analysis Tools, Data Collection and Analysis Tools, Evaluation and Decision Making Tools, Idea Creations Tools, Process Analysis Tools, Project Planning and Implementation Tools, Seven Basic Quality Tools, and Seven New Management and Planning Tools. Categories Description Tools Cause analysis tools Used to identify the cause of Fishbone diagram, Pareto chart, a problem. Scatter diagram Data collection and analysis tools Used to collect or analysis data. Evaluation and Used to select the best choices decision making tools or to evaluate what is performance level of project so far. Idea creations tools Used to create ideas or organize ideas. Process analysis tools Used when an understanding of process flow is desired. Seven basic quality tools Seven new management and planning tools These tools are the most fundamental tools of quality control. Used to encourage innovation, communicate information and successful planning of key projects. Table 2. Classification of tools and techniques according to ASQ (Source: American Society for Quality Website) Check sheet, Control chart, Design of experiment, Histogram, Scatter diagram, Stratification, Survey Decision matrix, Multi-voting Affinity diagram, Benchmarking, Brainstorming, Nominal group technique Flowchart, Failure mode effect analysis, Mistake-proofing Cause and effect diagram/ Fishbone diagram, Check sheets, Control charts, Histogram, Pareto chart, Scatter diagram, Stratification Affinity diagram, Relation diagram, Tree diagram, Matrix diagram, Arrow diagram, Process decision program chart

7 Qualitative and Quantitative Analysis of Six Sigma in Service Organizations 253 b. Tool Matrix In Nancy R. Tague s The Quality Toolbox (1995), she developed a Tool Matrix that classifies quality tools according to what the tools can offer. It is quite similar to the categorization suggested by ASQ, but differs, as it encompasses more tools. Categories Ideas creation Process analysis Cause analysis Planning Evaluation Data collection and analysis Table 3. Tool Matrix (Tague, 1995) c. Innovation Tools Tools Affinity diagram, Brainstorming. Brain writing Nominal group technique, Relation diagram Cost of quality analysis, Critical-to-quality analysis, Deployment flowchart, Flowchart Matrix diagram, Relations diagram, Requirements matrix, Requirements-and-measure matrix, Storyboard, Top-down flowchart, Work-flow diagram Contingency diagram, Fishbone diagram, Force field diagram, Is-is not matrix, Matrix diagram Pareto chart, Scatter diagram, Stratification, Tree diagram, Why-why diagram Activity chart, Arrow diagram, Contingency diagram, Deployment flowchart, Flowchart Force field analysis, Matrix analysis, Mission statement, Operational definitions, Plan-do-check-act cycle, Relations diagram, Storyboard, Top-down flowchart, Tree diagram, Work-flow diagram ACORN test, Continuum of team goals, Decision matrix, Effectiveachievable matrix, List reduction, Matrix diagram, Mission statement checklist, Multi-voting, Plan-results matrix, PMI Box plot, Check sheet, Control chart, Histograms, Importanceperformance analysis, Kologorov-Smirnov test, Normal probability plot, Operational definitions, Pareto chart, Performance index, Process capability, Requirements-and-measures tree, Run chart, Scatter diagram, Stratification, Survey The literature on service design and development talks about various tools which are effective in describing and analysing service problems. The tools are shown in Table 4. S. No. Tool Description 1 Structured analysis and Used to model service system design technique 2 Function analysis Maps customer requirements to required functions and means 3 Service blueprinting Analyse and represents the steps in a service process

8 254 Total Quality Management and Six Sigma 4 Quality function deployment (QFD) Translate customers needs and expectations into specifications that are relevant to companies 5 Root cause analysis Identify potential service failure points, service outcome or process problems in service recovery process 6 Theory of Inventive Problem Solving (TRIZ) Algorithmic approach for solving technical and technological problems 7 Axiomatic design Maintain the independence of the functional requirements and minimize the information content in a design Table 4. Innovation tools Six sigma methodologies DMAIC methodology Much information is available about the DMAIC (define, measure, analyze, improve, control) methodology. DMAIC is used mostly for existing processes. This approach not only makes use of various tools and techniques, it also incorporates other concepts such as financial analysis and project schedule development. The DMAIC methodology is excellent when dealing with an existing process in which reaching a defined level of performance will result in the benefits expected. There are number of articles and books providing details about DMAIC methodology. Table 5 provides the details about each phase taken from one of the literature. Phase Define Measure Analyse Improve Control Description Identify, evaluate and select projects; prepare the mission; and select and launch the team Measure the size of the problem, document the process, identify key customer requirements, determine key product characteristics and process parameters, document potential failure modes and effects; theorize on the cause or determinants of performance Plan for data collection; analyse the data and establish and confirm the vital few determinants of performance Design and carry out experiments to determine the mathematical causeeffect relationships and optimize the process Design controls; make improvements, implement and monitor Table 5. DMAIC methodology (Chua, 2001) Design for Six Sigma (DFSS): Overview The emergence of Six Sigma since 1980s has been phenomenal. Initially, the major focus of the organizations was to improve from their existing three sigma limits to Six Sigma limit of product or service quality. The importance of innovation in products and services has

9 Qualitative and Quantitative Analysis of Six Sigma in Service Organizations 255 changed the focus of organizations now more towards proactive approach rather than being reactive. The design for Six Sigma (DFSS) approach is relatively new compared to Six Sigma and is discussed in different ways in various literatures. Most of the literatures though agree that DFSS is a proactive approach and focuses on design by doing things right the first time. DFSS can be said as A disciplined and rigorous approach to design that ensures that new designs meet customer requirements at launch (El-Haik and Roy, 2005, pp 33). According to GE corporate research and development, the importance of DFSS is in the prediction of design quality up front and driving quality measurement and predictability improvement during the early design phases (Treichler et al, 2002). DFSS can also be explained as a datadriven methodology based on analytical tools which provide users with the ability to prevent and predict defects in the design of a product or service (De Feo and Bar-El, 2002). The major focus of DFSS approach is to look for inventive ways to satisfy and exceed the customer requirements. This can be achieved through optimization of product or service design function and then verifying that the product or service meets the requirements specified by the customer (Antony and Coronado, 2002). The literatures also concentrate on the differences between DMAIC and DFSS approach. Though DFSS involves designing processes to reach Six Sigma levels and is considered as an aggressive approach, but it still lacks a single methodology unlike Six Sigma (Hoerl, 2004). The different methodologies used in DFSS are: IDOV (Identify, Design, Optimize, Validate) ICOV (Identify, Characterize, Optimize, Validate) DCOV (Define, Characterize, Optimize, Verify) DMADO (Define, Measure, Analyze, Design, Optimize) DMADV (Define, Measure, Analyze, Design, Verify) DMADOV (Define, Measure, Analyze, Design, Optimize, Verify) DCCDI (Define, Customer Concept, Design, Implement) DMEDI (Define, Measure, Explore, Develop, Implement) Some of the other differences are: DFSS is a methodology that takes into account the issues highlighted by the end customers at the design stage while DMAIC solves operational issues (Ferryanto, 2005). Benefits in DFSS are difficult to quantify and are obtained in long term in comparison to Six Sigma, where the benefits are expressed mainly in financial terms and obtained rather quickly ( The DMAIC methodology tends to provide incremental improvements in comparison to DFSS where there can be radical improvements (El-Haik and Roy, 2005). The projects improved through DMAIC methodology are constrained by the assumptions made during the development and design stages, whereas DFSS builds quality into the design by implementing preventive thinking and tools in the product development process (Smith, 2001). The tools and techniques involved in the DFSS methodology are also somewhat different from those of the DMAIC methodology. DFSS includes innovation tools such as the theory

10 256 Total Quality Management and Six Sigma of inventive problem solving, axiomatic design, and quality function deployment, which DMAIC does not. Detailed information about the methodologies can be found in (Kwak and Anbari, 2006; Hendry and Nonthaleerak, 2005; El-Haik and Roy, 2005; Goel, et al., 2005; Raisinghani et al., 2005; Basu, 2004; Antony and Coronado, 2002; Stamatis, 2002 (a and b); Harry and Schroeder, 1999). Though there are differences among Six Sigma and DFSS approaches but still these two complement each other. Different DFSS stages are shown in Figure 1. Problem definition is the first stage, where customer requirements are incorporated. This stage is followed by the characterization stage. The model of the problem in the process or engineering domain is developed at this stage, which is basically the translation of the voice of customer and the customer usage conditions into an engineering system (Ferryanto, 2005). As seen from Figure 1, improvements from the DMAIC are added to the model at the characterization stage. After model development, optimal and robust solutions are found out. At the last stage the solutions are verified for their usefulness to solve the real problem. DFSS (Prevent defects in product development) Problems or Issues Solutions Six Sigma (Eliminate defects in current production) Define (Customer Domain) Characterize (Engineering Domain and Design of Experiments) Direction to improve manufacturing capability Measure (Production Domain) Analyze (Problem Solving) Optimize (Robust Design) Verify (Design Validation) Need to reduce Product sensitivity to manufacturing noise Improve (Elimination of Causes) Control (Statistical Process Control) Robustness Cost Savings Figure 1. DFSS versus Six Sigma (Ferryanto, 2005) * The DFSS model illustrated is Ford Motor Co. s DCOV approach

11 Qualitative and Quantitative Analysis of Six Sigma in Service Organizations Critical Success Factors (CSFs) CSFs are the essential ingredients required for success of Six Sigma projects in an organization (Coronado and Antony, 2002). There have been many studies on CSFs. One of the earliest is by Harry (2000), who discussed about six success factors involving management s leadership, belt system, etc. Later on Antony and Banuelas (2002) mentioned twelve success factors which include management involvement and commitment, linking Six Sigma to business strategy, etc. There are several other studies and all of them have at least one common CSF, i.e. top management commitment. The discussion on CSFs by Antony (2006) is the only one specific to service organizations. Some of the common CSFs are discussed below. i. Top management commitment and involvement Almost all the literature reviewed agrees that this factor is a must for successful Six Sigma implementation. And this has to be top-down rather than initiated by a particular department or from the ground (Goh, 2002). Top management involvement helps to influence and restructure the business organization and the cultural change in attitudes of individual employees toward quality in a short implementation period (Henderson and Evans, 2000). ii. Education and training Another important feature of Six Sigma is the elaborate training and certification processes that result in Black Belts, Green Belts, etc (Goh, 2002). Education and training help people understand the fundamentals of Six Sigma along with the application of tools and techniques to different phases of DMAIC. Training is part of the communication process to make sure that manager and employees apply and implement the Six Sigma techniques effectively (Kwak and Anbari, 2006). iii. Cultural change Six Sigma is considered a breakthrough management strategy, and it involves the adjustment of a firm s values and culture. In some cases, substantial change to an organization s structure and infrastructure need to take place (Coronado and Antony, 2002). People facing cultural change and challenges due to the implementation of Six Sigma need to understand this requirement. Also needed are a clear communication plan and channels to motivate individuals to overcome resistance and to educate senior managers, employees, and customers on the benefits of Six Sigma (Kwak and Anbari, 2006). iv. Customer focus Customer focus is one of the major requirements in implementing Six Sigma. This is emphasized in terms of critical to quality characteristics. Six Sigma is highly much more sensitive to requirements for customer satisfaction (Goh, 2002). v. Clear performance metrics This is an important factor from a service point of view. Often the difficulty is with identifying what to measure (Sehwall and De Yong, 2003). Before starting any Six Sigma initiative it is better to have a clear idea and agreement on the performance metrics to be used.

12 258 Total Quality Management and Six Sigma vi. Attaching the success to financial benefits Representing the success of Six Sigma projects in terms of financial benefits and measurement performance has made their selection and completion an important aspect for the organizations (Henderson and Evans, 2000). Financial benefits as a measure of achievement makes it easily understandable for the employees and helps them to relate to Six Sigma project outcome (Goh, 2002). vii. Organizational understanding of work processes The amount of effort that a service organization puts into measuring its work processes is important. Some organizations expend much time and effort in developing ways to measure the processes that ultimately impact customer satisfaction. Other organizations attempt this half-heartedly and measure only part of what is important to the customer. Like in hospitals the focus may be only on a particular laboratory or facility where the interaction with customer tends to be relatively greater. Because Six Sigma programs rely on measurements from processes, organizations with robust measurement systems in place are more likely to be ready for a Six Sigma implementation (Hensley and Dobie, 2005). The factors discussed above are equally applicable to services and manufacturing. Our literature review found that top management commitment, education and training, cultural change, and financial benefits are the most important CSFs. Figure 2 summarizes the importance of the CSFs as seen by each of the articles that were reviewed. Figure 2. Percentage of articles mentioning each of 19 CSFs 2.4. Critical-to-Quality (CTQ) characteristics In case of CTQs, we focused on its definitions mentioned in the literature. CTQ is defined in different ways in the literature but mostly they agree that it is a quality characteristic of product or service which is required to be improved from customer point of view. In other

13 Qualitative and Quantitative Analysis of Six Sigma in Service Organizations 259 words, CTQ is generated from critical customer requirements derived from voice of customer (refer Figure 3). Output Indicator CTQs Critical Customer Requirements Voice of Customer Figure 3. Understanding critical-to-quality (adapted from Muir, 2006) CTQs are the key measurable indicators of a product or process whose performance standards or specification limits must be met in order to satisfy the customer. CTQs align improvement or design efforts with customer requirements. In a layman term, CTQs are what customers expect of a product or service. They are the spoken needs of the customer (isixsigma/dictionary). Six Sigma focuses on process improvement, and improving the service process is a major determinant of customer satisfaction. The discussion on CTQs in the Six Sigma service literature is limited. Although services are widely different, the analysis from various literatures (Kwak and Anbari, 2006; Jones Jr., 2004; Sehwall and De Yong, 2003; Rucker, 2000) shows that some common CTQs exist across service. They are discussed below. i. Time (service time, waiting time, cycle time) In the case of services where the customer is involved in the process itself, time is an important consideration. The following three types of time should be considered: a. Service time: The time required to serve a particular customer b. Waiting time: The time customer waits in the system to get the work done c. Cycle time: The total time including service and waiting time. ii. Cost Like time, cost is sometimes a critical factor from the customer s point of view. The two are in fact intertwined. Customers may at times be willing to pay more for a service that can be completed in a shorter time. The trade-off between cost and time is, thus, important for services. iii. Employee behaviour For services where there is high degree of customer contact, employee behavior may be an important consideration. An employee s attitude towards a customer s problem may well decide whether the customer wishes to continue being serviced by the organization. iv. Information (accurate information, timely information) The growing importance of call center services shows the emergence of information needs. Getting the right information at the right time to one s customers is, thus, an important aspect from a customer point of view.

14 260 Total Quality Management and Six Sigma 2.5. Key Performance Indicators (KPIs) KPI is not well defined in the literature and there exist different interpretations of this term. Mostly the literature discuss about it as performance metrics, i.e., it is a measure of performance in terms of cost, quality, yield, and capacity (Basu and Wright, 2003; Hahn et al., 1999). A few of the suggested definitions of KPI is provided below (refer Table 6). Author(s) Hahn et al. (1999) Basu and Wright (2003) Antony (2006) ASQ Glossary KPI Definitions Performance metrics are established that directly measure the improvement in cost, quality, yield, and capacity. KPIs are measurements of a performance such as asset utilization, customer satisfaction, cycle time from order to delivery, inventory turnover, operations costs, productivity, and financial results. KPIs can be termed as performance metrics of Six Sigma. KPI is a statistical measure of how well an organization is doing in a particular area. A KPI could measure a company s financial performance or how it is holding up against customer requirements. Table 6. Key performance indicator definitions KPIs show actual data of a particular outcome. The outcomes of Six Sigma projects are usually required to be expressed in financial terms. This leads to a direct measure of achievement which is easy to understand (Goh, 2002). The majority of the KPI literature on Six Sigma in services talks about financial benefits. Other KPIs include expressions in terms of customer satisfaction and efficiency. Similar to CTQs, some KPIs are common across services. Some of the common KPIs are discussed below. i. Efficiency Efficiency in a service industry means the timely delivery of services at a reasonable cost. ii. Cost reduction Cost can be reduced by eliminating waste, such as reducing errors or mistakes in a process or reducing the time taken to complete a task. A concrete example is to reduce a patient s stay at a hospital (Heuvel et al., 2005), which can provide opportunity for more admissions. iii. Time-to-deliver Like in manufacturing, the time to deliver a service determines organizational performance. Examples may be the timely delivery of information or document as per customer requirement. iv. Quality of the service Quality of the service is a measure of the extent to which the service delivered, meets the customer s expectations. This depends on two aspects; one is the technical aspect and another is functional aspect. The technical aspect is the actual outcome of the service encounter. Functional aspect is the interaction between the service provider and customer i.e. the service process (Ghobadian et al., 1994).

15 Qualitative and Quantitative Analysis of Six Sigma in Service Organizations 261 v. Customer satisfaction This factor is difficult to measure as it varies from service to service. For example, for a call center service, customer satisfaction is measured by the receipt of timely information. For a hospital, the comfort and assurance that patient feels may be the all important criterion (Sehwall and De Yong, 2003). Overall customer satisfaction can also be indicated by the retention rate of one s customer. vi. Employee satisfaction This is another intangible measure of organizational performance. Employee retention rate can be an excellent indicator of employee satisfaction. Financial benefits due to Six Sigma can provide employees with a means to visualize their contribution. This may increase employee morale and satisfaction (Henderson and Evans, 2000). vii. Reduced variation Statistical process control and Six Sigma refer to the reduction of variation through improved standards and consistency. In the case of services, variation reduction may be in terms of, for example, the cycle time of processing statements, or the decision cycle of a process (such as credit process in a bank) or the inaccuracy of a billing process and incorrect laboratory test results (such as in a hospital) (Sehwall and De Yong, 2003; Rucker, 2000). viii. Financial benefits The impact of Six Sigma on the bottom line is huge (Henderson and Evans, 2000). In comparison to success and failure as a measure, financial bottom lines are a better indicator of the impact of improvements as well as a vivid calibration of progress (Goh, 2002) Six sigma in manufacturing and service organizations Although different terms may be used, scrap and rework exist in services just as they do in manufacturing. Inconsistent and out-of-specification processes cost money to rework. Such examples in services may include the need to re-contact a customer to verify an order, providing an incorrect service, providing a substandard service, or even over-servicing or providing more than what is required. Some widely publicized success stories due to implementation in services include GE Medical Systems, Mount Carmel Health System, Virtua Health, GE Capital Corp, Bank of America, and Citibank. Limited application can also be found in call centers, human resources such as DuPont de Nemours (Bott et al., 2000; Wyper and Harrison, 2000) and in product support services such as by Caterpillar (Schmidt and Aschkenase, 2004). The literature analysis also revealed that applications are limited mostly to service organizations in North America and Europe. Benefits-wise, these are mostly expressed in financial terms and not much is published about the benefits in process improvement terms. The literatures (Brady and Allen, 2006; Inozu et al., 2006; Mortimer, 2006; Antony et al., 2005a; Dudman, 2005; Goel et al., 2005; Hensley and Dobie, 2005; Basu, 2004; McAdam and Evans, 2004; Schimdt and Aschkenase, 2004; Hill and Kearney, 2003; Sehwall and De Yong, 2003; Rucker, 2000; Hahn et

16 262 Total Quality Management and Six Sigma al., 1999; Harry and Schroeder, 1999; Paul, 1999) on Six Sigma application in manufacturing or services discuss mainly about CSFs, CTQs, KPIs and STTs. The following section provides an overview of these factors. Table 7 presents the similarities and differences of these between manufacturing and services on the basis of observations from the literatures. CTQs STTs Dimensions Manufacturing Service CSFs Similarities Differences KPIs Similarities Differences Top management commitment, education and training, cultural change, linking Six Sigma to customers, linking Six Sigma to business strategy, effective communication Top management commitment, cultural change, clear performance metrics, customer focus, education and training, attaching the success to financial benefits, organizational understanding of work processes Cycle time, cost of quality, machine or human error Product performance Service time, waiting time, characteristics such as, employee behaviour, strength; weight, defects, poor responding to customer packaging, breakage, defects, complaints, providing inventory reduction, product accurate and timely travel distance, poor information to customers packaging, quantity of rework, time spent in rework Cost savings, customer satisfaction, reducing variation, employee satisfaction, increasing productivity, product quality improvement Efficiency, cost reduction, time to deliver, quality of the service, customer satisfaction, employee satisfaction Histogram, Pareto analysis, cause and effect analysis, brainstorming, flowchart, project charter, process mapping, root cause analysis, control charts FMEA, DOE, SPC, gauge repeatability and reproducibility, measurement system analysis, regression analysis, QFD Table 7. CSFs, CTQs, KPIs, and STTs (manufacturing versus service) The above table provides some important insights regarding Six Sigma implementation aspects in manufacturing and services. There are similar CSFs in manufacturing and services but their order of preference differs between two. This difference in order of preference can also be observed within the literature involving Six Sigma implementation in services. The paper by Antony (2004b) shows that linking Six Sigma to business strategy is the most important of success factors whereas some other literatures discuss that top management commitment is the most important one, followed by education and training (Johnson and Swisher, 2003; Henderson and Evans, 2000).

17 Qualitative and Quantitative Analysis of Six Sigma in Service Organizations 263 CTQs show similarities in terms of cycle time and cost. The concentration in manufacturing is more on product specifications/characteristics, inventory reduction, and reducing variation whereas services focus more on service time, waiting time, responding to customer, employee behaviour, etc. The reason for this difference can be because of more customer contact in services. KPIs for both manufacturing and services show much similarity and are not much discussed in literatures. The application of tools and techniques has similarities in usage of flowcharts, process map, histograms, Pareto analysis, etc. The use of statistical tools and techniques such as SPC and regression analysis is more prominent in manufacturing may be because of ease of data collection and continuity of the process. The tools and techniques such as gauge repeatability and reproducibility is commonly used in manufacturing but not so in services, the reason is non-repeatable nature of service processes (Does et al., 2002) Review summary First, although the industry has an increased interest in Six Sigma implementation and many companies have gained the profits and advantages from this disciplined approach, the literature is limited and the research impacts of Six Sigma implementation and factors contributing to its success remain unclear. Many articles on the impact analysis of operations performance do not mention the detailed improvements in the operating areas, but focus on the overall bottom line impact. Therefore, it is necessary to do a deeper and more detailed study in this area. Second, only a few articles were found that dealt with factors in the area of success factor analysis to Six Sigma implementation. Existing studies are not well integrated and current concepts in the field of Six Sigma are largely based on case studies, anecdotal evidences and are prescriptive in nature. Consequently there is little consensus on which factors are critical to the success of the approach (Nonthaleerak and Hendry, 2008; Brady, 2005). Most of the articles concentrated on few success factors and reported that top management commitment is the main factor to Six Sigma success (Goh, 2002; Henderson and Evans, 2000). However, many other factors affecting Six Sigma s success are important and need to be better documented. To fill this gap, Antony and Baneulas (2002) identified 10 typical CSFs from their review of literature. Several others also provided sets of CSFs which have similarities or differences among them. It could be argued that this list of CSFs is comprehensive and that many of the issues are in common with those found for any implementation process, and are thus not specific to Six Sigma. However, all of the papers that identify these issues are descriptive in nature and there is a need to verify them through rigorous empirical research. Finally, some authors have called for theoretic research (Nonthaleerak and Hendry, 2008; Schroeder et al., 2008; Oke, 2007; Brady and Allen, 2006), as too much research is focused only on description of practice rather than on theory development that is of use to practitioners as well as academics.

18 264 Total Quality Management and Six Sigma 3. Research methodology Management research is mainly based on deductive theory testing and positivistic research methodologies (Alvesson and Willmott, 1996). These methodologies incorporate a more scientific approach with the formulation of theories and the use of large data samples to observe their validity. However, these approaches mostly fail to give deep insights and rich data in Six Sigma practice within service organizations. Schroeder et al. (2007) state the need for more theory grounded and contingency based research rather than be restrictive to deductive approaches. Antony et al. (2007) and Nonthaleerak and Hendry (2008) emphasize this point by saying there is a paucity of systematic and rigorous evaluation in many Six Sigma studies. In this section we describe the three phase approach for this study. First phase involved literature review and exploratory case studies. A small-scale questionnaire survey and 15 case studies were done in the second phase. The third phase included a large-scale questionnaire survey and further case studies. Questionnaire structure and design for each phase is also discussed. Then details about the measures are provided. Finally, we explain how we test the sample bias, which population is targeted, and how to proceed for the data collection Phase I Macro study This phase focused on providing the necessary breadth to produce an understanding of the implementation of Six Sigma in service organizations and from which reliable patterns and theories can be formed. Next phase of this research focused on the issues uncovered by the first. The phase is termed as macro study (Leonard and McAdam, 2001), and it provides an overview not only of Six Sigma implementation in services, but also a database of critical success factors (CSFs), critical-to-quality (CTQ) characteristics, key performance indicators (KPIs), and set of tools and techniques (STTs). The study included two services one is library and the other one is a call center. During this phase, interviews were conducted with a black belt, who was considered by the organization as most knowledgeable and responsible for Six Sigma implementation. The study concentrated on the implementation aspect of Six Sigma which involves CSFs, CTQs, KPIs, STTs, and also the difficulties faced. The BB provided an essential insight and understanding of Six Sigma implementation in service organizations. The other methods of data collection in this phase involved documentation and archival records. Once the macro study is completed and insights are developed, preparation for next phase was done to focus on additional relevant questions that had arisen in phase one. This next phase involved a small-scale questionnaire survey and simultaneous case studies. The study included multiple respondents which overcame the problem of using single respondent in phase one. It also provided a degree of validation.

19 Qualitative and Quantitative Analysis of Six Sigma in Service Organizations Phase II Small-scale questionnaire survey and case studies Small-scale questionnaire survey At this phase a questionnaire survey of Singapore service organizations was conducted to understand the status of Six Sigma implementation. The survey was exploratory in nature as the objective was to gain insights about Six Sigma in service organizations. This kind of survey helps to uncover or provide preliminary evidence of association among concepts. Further, it can help to explore the valid boundary of a theory (Forza, 2002). Phase I Macro study Exploratory case studies Action Redefine/expand/refocus research agenda Develop new ideas/directions Phase II Companies in similar industry or in similar geographical location Small scale survey Conversations with practitioners Observation study Case Studies Outcomes Describe data to which the company could specifically respond with tangible actions Experiential knowledge development of researcher Ongoing interpretive analysis, allowing data to be initially coded in several ways, then reanalyzed and interpreted as further data are gathered Holistic perspective / understanding of phenomena Action Add wider dimensions to agenda Redefine/expand/refocus research agenda Develop new ideas/directions Phase III Companies in similar industry in different geographical locations/or specific aspects within one company Large scale survey In-depth interviews Conversations with practitioners Content analysis of organization materials Figure 4. Three-phase approach (adapted from Leonard and McAdam, 2001; Gilmore and Carson, 1996)

20 266 Total Quality Management and Six Sigma Structure of the questionnaire There are five parts in the questionnaire. The first part of the questionnaire is intended to get some general information of the respondent company, which includes the type of service organization, the size of the company, the type of company (local, multinational, or joint venture), whether they have quality department, there is a proper quality system in place, any business process improvement initiatives they are doing, and finally whether they have implemented Six Sigma. It is also designed as a filter to segregate the data based on service organizations which have or have not implemented Six Sigma. The second part of the questionnaire attempts to identify the CSFs which are important while implementing Six Sigma in organization. The third part consists of two questions. First question is directed at identifying CTQs that are to be improved through Six Sigma implementation. Second question explores the tools and techniques used in Six Sigma DMAIC methodology and also in DFSS. The fourth part is focused on finding about KPIs while in the fifth part the objective is to identify the difficulties faced by service organizations in Six Sigma implementation. The fifth part is for those service organizations which have not implemented Six Sigma. There is one question in this part to explore about the reasons behind not implementing Six Sigma. The last part is designed to obtain background information on respondents including their name, job title, company, mailing address, phone/fax number, and . In order to share our findings with the respondents who are interested, we also left a space for them to tick whether they want to have the summary of our survey results. Besides the six parts above, a cover letter with university letterhead explaining the aims and benefits of the research was designed. Questionnaire design The response format of the questionnaire is a major design consideration since this will alter the type and wording of the questions as well as focus on the type of analysis that the researcher wants to perform (Antony et al., 2007; Fowler, 2002; Kidder, 1986). For our research close-ended question format was considered since the data would be in a quantifiable form ensuring that statistical analysis can be used. Moreover, it is fast and easy to complete, enables automated data entry, and facilitates data analysis and summary of data (Antony et al., 2007; Fowler, 2002). The rating scale (Likert scale) and ranking used within this format is to obtain the answers from the respondents. The questionnaire focused on CSFs, CTQs, STTs, and KPIs as observed from the literature. CSFs, STTs, and KPIs are measured by a 5-point Likert-type scale (CSFs and KPIs: 1 = not important, 5 = very important; STTs: 1 = never, 5 = frequently). The Likert scale used provide a more precise measure than yes/no or true/false items and it is fast and easy to complete (Neuman, 2006). The rating scale used for few questions allows the respondents to indicate the relative importance of choices that facilitates the researchers in identifying the critical issues or factors (Antony et al., 2007).